> ## Documentation Index
> Fetch the complete documentation index at: https://docs.geminix.cc/llms.txt
> Use this file to discover all available pages before exploring further.

# Configure Base URL

> Set the OmniMux API Base URL in clients, SDKs, and CLIs.

## Production endpoints

| Purpose                               | URL                                                          |
| ------------------------------------- | ------------------------------------------------------------ |
| Console                               | [https://geminix.cc/dashboard](https://geminix.cc/dashboard) |
| API Base URL                          | `https://api.geminix.cc`                                     |
| OpenAI-compatible (typical SDK value) | `https://api.geminix.cc/v1`                                  |
| Docs                                  | [https://docs.geminix.cc](https://docs.geminix.cc)           |

Auth header:

```http theme={null}
Authorization: Bearer sk-xxxxxxxx
```

## Which value to paste?

| Context                                       | Value                                        |
| --------------------------------------------- | -------------------------------------------- |
| Official OpenAI SDK / most compatible clients | `https://api.geminix.cc/v1`                  |
| Full path you build yourself (curl)           | `https://api.geminix.cc/v1/chat/completions` |
| “Gateway root only”                           | `https://api.geminix.cc`                     |

**Do not** use the console host `https://geminix.cc` as the API Base URL when the API is served on `api.geminix.cc`.

## Common clients

### 1) Environment variables

```bash theme={null}
export OMNIMUX_API_KEY="sk-xxxxxxxx"
export OPENAI_API_KEY="$OMNIMUX_API_KEY"
export OPENAI_BASE_URL="https://api.geminix.cc/v1"
```

### 2) Python (openai)

```python theme={null}
from openai import OpenAI

client = OpenAI(
    api_key="sk-xxxxxxxx",
    base_url="https://api.geminix.cc/v1",
)
```

### 3) Node.js (openai)

```javascript theme={null}
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.OMNIMUX_API_KEY,
  baseURL: "https://api.geminix.cc/v1",
});
```

### 4) curl

```bash theme={null}
curl https://api.geminix.cc/v1/models \
  -H "Authorization: Bearer $OMNIMUX_API_KEY"
```

### 5) Cherry Studio / OpenAI-compatible GUIs

1. Settings → Model provider / API
2. **API base / Base URL**: `https://api.geminix.cc/v1`
3. **API Key**: token from the console (`sk-...`)
4. Save, then list models or send a chat

### 6) Cursor / Continue

In the provider config:

* Base URL: `https://api.geminix.cc/v1`
* API Key: your token

Field labels vary (`OpenAI Base URL`, `Override Base URL`); the value is the same.

## Self-check

```bash theme={null}
curl -sS https://api.geminix.cc/api/status | head -c 200

curl -sS https://api.geminix.cc/v1/models \
  -H "Authorization: Bearer $OMNIMUX_API_KEY"
```

| Symptom                  | Fix                                                            |
| ------------------------ | -------------------------------------------------------------- |
| DNS / not resolved       | Add CNAME for `api` (see DNS ops)                              |
| 401                      | Bad key or missing `Authorization`                             |
| 404 on `/v1/...`         | Extra or missing `/v1` relative to how the client builds paths |
| Console works, API fails | Console is `geminix.cc`; API should be `api.geminix.cc`        |

## Related

* [Quickstart](/en/quickstart)
* [Authentication](/en/guides/authentication)
* [Models](/en/guides/models)